The Price of Happy Hens: A Hedonic Analysis of Retail Egg Prices

Page created by Dennis Elliott
 
CONTINUE READING
Journal of Agricultural and Resource Economics 35(3):406–423
Copyright 2010 Western Agricultural Economics Association

                       The Price of Happy Hens:
                  A Hedonic Analysis of Retail Egg Prices
                  Jae Bong Chang, Jayson L. Lusk, and F. Bailey Norwood

           This paper analyzes price differentials among conventional, cage-free, organic, and
           Omega-3 eggs using retail scanner data from two regional markets and the United States
           as a whole. Results reveal significant premiums attributable to cage-free (a 57% premium
           on average) and organic (an 85% premium on average). However, significant variation
           exists among geographic locations; price premiums for organic over conventional eggs in
           Dallas are almost twice as high as those in San Francisco. Estimates indicate that about
           42% of the typically observed premium for cage-free eggs over conventional eggs (and
           36% of the premium for organic eggs) can be attributed to egg color rather than differ-
           ences in hens’ living conditions. Despite the large implicit price premiums for cage-free
           and organic, our data reveal that most shoppers are not willing to pay such high prices for
           cage-free and organic attributes.

           Key Words: animal welfare, cage-free, eggs, free-range, hedonic, organic

                                                     Introduction

The vast majority of commercial egg-producing hens in the United States live in battery cage
systems. A typical egg farm contains anywhere from 100,000 to 1 million hens, with each hen
housed in a metal wire cage with four or five other birds. These production systems have been
criticized by animal advocacy groups for several reasons: the tight space requirements (about
67 square inches of space per hen) prohibit chickens from fully extending their wings, and
hens are unable to exhibit natural behaviors such as dust bathing and laying eggs in nests.
These concerns recently received widespread public attention when the Humane Society of
the United States (HSUS), working with several other advocacy groups, introduced a ballot
initiative to ban battery cage egg production in the state of California. The proposition,
formally known as the Prevention of Farm Animal Cruelty Act, passed in November 2008,
with 63.5% of Californians voting in favor. The effects of the ban in California have been
intensely debated, and one of the key unanswered questions is the potential impact of the ban
on egg prices.
    In addition to policy initiatives, increasing consumer demand for animal welfare, human
health, and variety has prompted producers to market a diversity of egg offerings. Demand for
health attributes has resulted in the development of Omega-3 and vitamin-enriched eggs, and
labeling programs facilitated by the U.S. Department of Agriculture, the United Egg Producers,
and the Animal Welfare Institute have led to cage-free, organic, and free-range egg options.
Although sales of generic eggs have decreased over the past decade, specialty egg sales have

Jae Bong Chang is a research fellow at the Korean Rural Economics Institute; Jayson L. Lusk is a professor and Willard Sparks
Endowed Chair; and F. Bailey Norwood is an associate professor, all in the Department of Agricultural Economics, Oklahoma
State University. The authors would like to thank three anonymous reviewers and the editor for helpful comments.
  Review coordinated by Gary W. Brester.
Chang, Lusk, and Norwood                               A Hedonic Analysis of Retail Egg Prices 407

steadily increased, accounting for approximately 16% of total egg sales in 2005. Organic eggs
represent one of the fastest growing segments in this market; sales grew at an average annual
rate of 19% from 2000 to 2005, reaching $161 million in 2005 (Oberholtzer, Green, and
Lopez, 2006). As a result, the egg market is now highly differentiated, and most grocery
stores currently offer eggs differentiated by production method, color, size, brand, and pack-
aging.
   Despite these changes, there is presently little publicly available information on price
differences between various types of eggs or on consumers’ willingness to pay for animal
welfare and health attributes. To address this issue, we carry out a hedonic analysis using
retail scanner data for two regional markets and for the United States as a whole.

                                        Background

Egg production is a large and highly concentrated business. Each year, more than 300 million
laying hens produce over 90 billion eggs in the United States. In 2007, five states (Iowa,
Ohio, Indiana, Pennsylvania, and Texas) produced 40% of eggs. The industry is moderately
concentrated, with 205 egg-producing companies owning 95% of all layers in the United
States (American Egg Board, 2009).
   Egg consumption fell dramatically from the 1940s to the 1990s due to concerns about
health and cholesterol; annual per capita egg consumption peaked at 403 in 1945, and reached
a low of about 230 in the 1990s (Putnam and Allshouse, 1999). Per capita consumption of
eggs has subsequently stabilized and was approximately 243 eggs per person per year in 2008
[USDA/Economic Research Service (ERS), 2009]. As consumer demand for eggs has changed
and the market has become increasingly differentiated, several studies have investigated
consumer preferences for egg characteristics. These studies relied on surveys to assess
Canadian and UK consumer egg preferences. Fearne and Lavelle (1996) found that price and
animal welfare were two key factors driving egg choice for consumers in the United
Kingdom. Bennett (1997) used data collected from mail surveys to measure UK consumers’
willingness to pay to support legislation to ban the use of battery cages in egg production;
almost 80% of survey respondents supported the proposed legislation and were willing to
pay, on average, £0.43 more per dozen given the ban. Asselin (2005) studied Canadian
consumers’ willingness to pay for Omega-3 and vitamin-enhanced eggs, and found people
were willing to pay between $0 and $0.72 more per dozen for Omega-3 eggs, depending on
their health consciousness.
   A few studies have employed observed market data to evaluate consumer egg preferences.
Baltzer (2004) used scanner data on weekly sales from a major retail chain in Denmark. He
estimated an almost ideal demand system (AIDS) model to determine Danish consumers’
marginal willingness to pay for five varieties of eggs—battery cage, barn, free-range, organic,
and pasteurized eggs—and found consumers were willing to pay significant premiums for
eggs that improved animal welfare and involved natural production methods. Karipidis et al.
(2005) used data obtained from retail stores in two metropolitan areas in Greece during the
summer of 2004. In their hedonic price modeling, 14 egg attributes were clustered into five
groups: natural characteristics, poultry-feeding conditions, packaging, quality-control system,
and vertical integration. The authors reported that eggs produced under organic and free-
range feeding systems and Omega-3 enriched eggs have high hedonic prices, but other
feeding conditions did not influence retail prices. Their study, however, included only 175
observations and a limited set of attribute types and levels.
408 December 2010                                                      Journal of Agricultural and Resource Economics

                                                          Methods
Data

Our data consist of three sets of point-of-sale scanner data collected from a representative
sample of retail outlets by Information Resources, Inc. (IRI).1 Each set contains data on stock
keeping units (SKUs), including information on size, color, package type, brand vs. private
label, and the number of eggs per package. Moreover, the data set also indicates whether each
SKU was advertised as cage-free, cage-free & Omega-3, cage-free & organic, fertile, free-
range, free-range & organic, natural, Omega-3, organic, organic & Omega-3, pasteurized,
vegan fed, or generic fresh eggs (i.e., battery cage eggs).2 Color levels of eggs include white
and brown, and size levels are small, medium, large, extra large, jumbo, and super jumbo.
Packaging was categorized by IRI into eight types: carton, cardboard carton, plastic carton,
styrofoam carton, paper carton, plastic wrapped carton, cardboard box, and other type.
   The first data set consists of quarterly egg sales for over 1,985 SKUs across the entire
United States over the 20 quarters from the beginning of 2004 to the end of 2008. In addition
to the U.S. data, we also obtained weekly scanner data for two specific metropolitan areas:
Dallas/Fort Worth, Texas, and San Francisco/Oakland, California. These data sets span from
January 1, 2007 to January 25, 2009, and are in all respects identical to the overall U.S. data
set except for data frequency (weekly rather than quarterly) and time period covered (two
years starting in January 2007 vs. four years starting in January 2004). We chose these two
specific locations for several reasons. First, we wanted a California location because of the
debate over the recent California ballot initiative. Second, we wanted a location of a similar
size that was some distance from California (and thus less likely to be susceptible to media
exposure on the ballot initiative), and one that we perceived might be less interested in animal
welfare issues.3 In addition, Texas and California rank fifth and sixth among egg-producing
states in number of eggs produced.
   Table 1 reports the percentage of SKUs possessing the various attributes considered in this
analysis. About 85% are generic fresh eggs (either conventional fresh or with no production
method indicated), while the number of SKUs advertised as organic represents only 5.28% of
the total. Another 1.22 + 0.05 + 0.58 = 1.85% of SKUs advertised organic in addition to
another attribute such as cage-free, free-range, or Omega-3. Approximately 70% of egg SKUs
are white. Most SKUs are large-sized eggs (53%) and sold by the dozen (65%). A larger
percentage of the SKUs in San Francisco are cage-free and organic than is the case in
Dallas. Likewise, a larger percentage of SKUs are brown vs. white eggs in San Francisco as
compared to Dallas. About 33% of the SKUs in Dallas are associated with package sizes
including 13 or more eggs, but only about 20% of SKUs in San Francisco include this many
eggs.

   1
     IRI records weekly data from 34,000 supermarkets, drug stores, mass merchandisers, and convenience channels across the
United States. The U.S. data set contains information on approximately 108 billion egg sales from 2004 to 2008. USDA/ERS
(2009) data indicate that about 258 billion shell eggs were produced during this time period. Therefore, our data set covers
approximately 42% of all U.S. shell egg production from 2004 to 2008.
   2
     The data set also included sales of cooked, cultured, hard-boiled, and homogenized eggs. We excluded these types of eggs in
our analysis. We also removed SKUs from the data set in which package sizes were only one egg per package because the per egg
price of such SKUs was often an order of magnitude higher than the mean.
   3
     See Lusk (2010) for an analysis of how demand for cage-free and organic eggs differed and changed over time in these two
locations. Note also that the Humane Society of the United States (2010) recently released a comprehensive report, “Humane State
Ranking,” rating all 50 states on a wide range of animal protection laws dealing with pets, animal cruelty and fighting, wildlife,
animals in research, and farm animals. California is listed at the top and Texas comes in at 33rd place.
Chang, Lusk, and Norwood                            A Hedonic Analysis of Retail Egg Prices 409

Table 1. Percentage of Fresh Egg SKUs Possessing Various Attributes by Market
                                                       Percent of Total SKUs
 Attribute                          United States   San Francisco / Oakland    Dallas / Ft. Worth

 Production Method:
   Cage-Free                            1.98                  4.75                   2.92
   Cage-Free & Omega-3                  0.11                  0.60                    —
   Cage-Free & Organic                  1.22                  4.13                   3.87
   Fertile                              0.22                  2.39                    —
   Free-Range                           0.45                   —                      —
   Free-Range & Organic                 0.05                   —                      —
   Natural                              1.25                  1.19                   0.97
   Omega-3                              1.99                  1.09                   3.85
   Organic                              5.28                 10.02                   2.93
   Organic & Omega-3                    0.58                  1.31                   0.02
   Pasteurized                          0.72                  1.19                   0.97
   Vegan Fed                            0.80                  3.58                   1.03
   Conventional Fresh                  74.42                 61.48                  72.46
   None Indicated                      10.92                  8.26                  10.98
 Color:
   White                               70.16                 63.88                  76.07
   Brown                               25.13                 33.34                  20.95
   None Indicated                       4.70                  2.78                   2.98
 Size:
   Small                                2.55                  0.49                   1.32
   Medium                              13.24                  9.31                  14.46
   Large                               52.92                 59.74                  60.38
   Extra Large                         17.61                 22.38                  14.85
   Jumbo                               12.11                  8.09                   8.00
   Super Jumbo                          0.47                   —                      —
   None Indicated                       1.09                   —                     0.97
 Package:
   Carton                              38.94                 45.85                  44.43
   Cardboard Carton                    23.07                 38.08                  13.69
   Plastic Carton                       5.65                  3.41                   6.70
   Styrofoam Carton                    23.72                  9.42                  25.43
   Paper Carton                         1.05                  0.19                   0.97
   Plastic Wrapped Carton               1.08                  1.19                   1.64
   Cardboard Box                        1.10                  0.80                   0.65
   Others                               1.52                   —                     1.09
   None Indicated                       3.86                  1.07                   5.40
 Brand Label:
   Private                             18.57                 31.30                  56.29
   Specific                            81.43                 68.70                  42.71
 Number of Eggs per Package:
   3–6                                  8.17                 11.49                   5.48
   7–11                                 3.21                  1.48                   0.97
   12                                  64.94                 67.13                  60.45
   13–18                               12.26                  9.21                  15.02
   > 18                                11.42                 10.69                  18.08
 Number of Unique SKUs                 1,985                  122                    150
410 December 2010                                                  Journal of Agricultural and Resource Economics

Table 2. Percentage of Fresh Eggs Sold by Production Method and Market
                                                                        Percent of Total SKUs
 Attribute                                         United States     San Francisco / Oakland    Dallas / Ft. Worth

 Production Method:
   Cage-Free                                           1.17                    2.29                   0.50
   Cage-Free & Omega-3                                 0.04                    0.07                    —
   Cage-Free & Organic                                 0.30                    1.76                   0.21
   Fertile                                             0.06                    2.70                    —
   Free-Range                                          0.04                     —                      —
   Free-Range & Organic                                0.00                     —                      —
   Natural                                             0.49                    0.45                   0.06
   Omega-3                                             0.52                    0.36                   0.54
   Organic                                             1.02                    4.43                   0.41
   Organic & Omega-3                                   0.13                    0.12                   0.00
   Pasteurized                                         0.19                    0.06                   0.05
   Vegan Fed                                           1.26                    0.95                   1.33
   Conventional Fresh                                 87.35                   84.47                  79.40
   None Indicated                                      7.43                    2.34                  17.50
 Number of Eggs Sold (billion)                       108.076 a                0.7256 b              0.9408 b
 a
     Total eggs sold over the 20 quarters from 2004 to 2008.
 b
     Total eggs sold over the 108-week time period from January 1, 2007 to January 25, 2009.

   Although our hedonic analysis uses the price of an SKU as the unit of measure, it is also
instructive to note the volume of eggs actually sold; this information is relevant to discussions
about consumer demand for different types of eggs. To provide a feel for differences in sales
volume across different production systems, we used the scanner data to calculate the quantity-
share of eggs sold using each production method by location (table 2). In the United States,
almost 95% of eggs sold are conventional fresh or generic eggs (eggs with “none indicated”
for production method are almost certainly cage eggs). While about 5.28% of SKUs are
organic, only about 1% of egg sales are organic.
   Table 2 also reveals regional differences. Of all eggs sold in San Francisco, 4.43% and
2.29% are organic and cage-free, respectively. Dallas consumers bought fewer organic and
cage-free eggs—only 0.41% and 0.50%, respectively. Overall, “specialty” eggs account for
about 30% of total SKUs in San Francisco and 17% of SKUs in Dallas, but only about 13%
and 3% of total eggs sold in San Francisco and Dallas. The small market shares do not
necessarily imply a small volume of egg sales. The last row of table 2 shows that almost a
billion eggs were sold in the Dallas/Ft. Worth stores over the two-year time period, and the
average weekly volume of “specialty” egg sales was about 886,000 eggs per week in San
Francisco and 270,000 eggs per week in Dallas.
   For each SKU, we calculate the average price paid in each time period (each quarter for the
U.S. data and each week for the regional data sets). Egg prices are deflated by the Consumer
Price Index (2008 = 100). Because SKUs differ in the number of eggs per package, we
calculate prices on a per egg basis. Figure 1 illustrates the trend in U.S. egg prices ($/egg) for
three different production methods: conventional, cage-free, and organic. The price premium
for organic over conventional fresh eggs ranged from $0.16/egg in the first quarter of 2008 to
$0.21/egg during 2005. The average price premium for organic eggs over the entire period
was about $0.20/egg. The average price premium for cage-free over conventional eggs was
Chang, Lusk, and Norwood                                  A Hedonic Analysis of Retail Egg Prices 411

        $0.40

        $0.35

                                                Organic
        $0.30

        $0.25
                                                Cage-Free

        $0.20

        $0.15
                                                                Conventional

        $0.10

        Figure 1. Average U.S. egg price ($/egg) over time by production method

about $0.12/egg over the time period and reached a minimum of $0.08/egg during early 2008,
when conventional egg prices rose sharply. The price trends shown in figure 1 are calculated
based on the raw data (adjusted for inflation) and do not account for differences that are often
correlated with organic eggs. For example, organic eggs tend to be brown and tend to be sold
in smaller package sizes (although this is not exclusively true), whereas conventional fresh
eggs tend to be white and sold by the dozen. Thus, the raw price premia for organic vs.
conventional ($0.20/egg) is attributable to a variety of factors other than “organic,” a fact high-
lighting the need for hedonic regression analysis to calculate ceteris paribus effects.
   Table 3 reports descriptive statistics for egg prices. The data reveal that prices tend to be
higher in San Francisco as compared to Dallas or the rest of the country as a whole. For
example, the average price of conventional fresh eggs was $0.26/egg in San Francisco but
only $0.14/egg in Dallas. Many conditions contribute to the price difference between the two
locations. In particular, the overall cost of living and average wage rates are higher in San
Francisco than in Dallas. Differences in the costs of egg production and transportation are
also relevant. Somewhat surprisingly, the premiums for organic and cage-free eggs over
conventional eggs were lower in San Francisco ($0.13/egg and $0.06/egg, respectively)
compared to Dallas ($0.20/egg and $0.12/egg, respectively) or the rest of the country ($0.20/
egg and $0.12/egg, respectively).

The Hedonic Price Model

The primary objective of this paper is to estimate a descriptive model of egg prices to
determine how egg prices differ by product characteristics such as organic and cage-free. We
used the standard hedonic modeling approach to accomplish this task. The hedonic method
relies on observing differences in market prices to infer the value or implicit price
of underlying characteristics. The method has been used at least since Waugh (1928), and was
412 December 2010                                                  Journal of Agricultural and Resource Economics

Table 3. Egg Price ($/egg) by Location and Production Method
                                    United States             San Francisco / Oakland            Dallas / Fort Worth
                           Mean                               Mean                            Mean
Production Method       (Std. Dev.)     Min.        Max.   (Std. Dev.)   Min.     Max.     (Std. Dev.)    Min.         Max.

Cage-Free                   0.26         0.13       0.41      0.32       0.17     0.40        0.26        0.12         0.34
                           (0.05)                            (0.04)                          (0.04)
Cage-Free & Omega-3         0.22         0.18       0.27      0.36       0.34     0.38         —           —            —
                           (0.03)                            (0.01)
Cage-Free & Organic         0.33         0.13       0.61      0.41       0.23     0.51        0.37        0.31         0.47
                           (0.10)                            (0.07)                          (0.05)
Fertile                     0.28         0.22       0.35      0.29       0.21     0.32         —           —            —
                           (0.03)                            (0.02)
Free-Range                  0.43         0.16       1.23       —         —         —           —           —            —
                           (0.36)
Free-Range & Organic        0.30         0.18       0.35       —         —         —           —           —            —
                           (0.05)
Natural                     0.20         0.07       0.35      0.26       0.17     0.33        0.16        0.13         0.21
                           (0.05)                            (0.05)                          (0.03)
Omega-3                     0.27         0.07       0.61      0.32       0.17     0.40        0.22        0.08         0.27
                           (0.09)                            (0.05)                          (0.02)
Organic                     0.34         0.03       0.72      0.39       0.08     0.51        0.34        0.26         0.57
                           (0.10)                            (0.04)                          (0.02)
Organic & Omega-3           0.38         0.18       0.53      0.38       0.23     0.49        0.47        0.43         0.51
                           (0.07)                            (0.04)                          (0.06)
Pasteurized                 0.19         0.06       3.61      0.36       0.33     0.40        0.29        0.24         0.30
                           (0.34)                            (0.02)                          (0.01)
Vegan Fed                   0.24         0.13       0.33      0.33       0.21     0.37        0.24        0.21         0.26
                           (0.03)                            (0.02)                          (0.01)
Conventional Fresh          0.14         0.01       2.09      0.26       0.08     0.68        0.14        0.03         0.45
                           (0.07)                            (0.07)                          (0.05)
None Indicated              0.15         0.02       0.88      0.23       0.08     0.33        0.16        0.07         1.29
                           (0.07)                            (0.06)                          (0.09)

Notes: Egg prices in the U.S. are from the first quarter of 2004 to the fourth quarter of 2008, and egg prices in the San
Francisco / Oakland and Dallas / Fort Worth areas are from January 1, 2007 to January 25, 2009. Prices are deflated by the
consumer price index (2008 = 100).

formalized in Rosen’s (1974) seminal study showing how the “1st stage” hedonic price
function could be derived via the interaction of consumer demand for and producer supply of
product characteristics. The “1st stage” hedonic price functions we estimate show how egg
prices in equilibrium vary with egg characteristics, which relate to a point on the marginal
willingness-to-pay schedule. However, implicit prices do not reveal a full willingness-to-pay
or demand schedule because the marginal implicit price is also determined by the producers’
offer curve, which varies with the costs of providing different characteristics. Differences in
cost across location or difference in demand can therefore explain differences in estimated
implicit prices. Recent studies (e.g., Bajari and Benkard, 2005) have questioned the appropri-
ateness of some of the assumptions of Rosen’s model—such as the assumption of perfect
competition, separability of attributes, and continuous product space—insofar as they allow
the analyst to identify consumers’ preferences for the underlying product characteristics.
   If we make the standard assumption that any unobservables are uncorrelated with the
observed product characteristics, the empirical model can be written as:
Chang, Lusk, and Norwood                                                         A Hedonic Analysis of Retail Egg Prices 413

                                          13                                          2                      6
(1)           ln( Priceit )  0    1j Production Methodi j    2j Colori j    3j Sizei j
                                          j 1                                       j 1                    j 1
                                                                  8
                                   4 Private Labeli    5j Packageij  6 (# Eggs / Package) i
                                                                 j 1
                                                                          11
                                   7 (# Eggs / Package ) i2    8j Month i j  it ,
                                                                          j 1

where Priceit is the average price ($/egg) for SKU i in time period t, the β’s are parameters to
be estimated,  it is a stochastic error term, and the remaining variables are dummy variables
identifying the attributes shown in table 1, except for the variable #Eggs/Package, which is a
continuous variable indicating the number of eggs per package. As previously mentioned, the
U.S. data set consists of quarterly observations, and the two regional markets consist of weekly
observations, so monthly dummy variables in equation (1) are replaced with three quarterly
dummy variables when analyzing the national data set.
   Potential options for functional forms are restricted given the large number of dummy vari-
ables in our model (McConnell and Strand, 2000; Oczkowski, 1994). As shown in equation
(1), we chose the semi-log model over the linear model based on the results from Box-Cox
specification tests and based on previous literature arguing that errors in the semi-log model
are relatively more homoskedastic compared to errors in the linear model (Diewert, 2003).
When dummy variables are included in a semi-log model, the resulting coefficients are not
equal to percentage changes (Halvorsen and Palmquist, 1980). For the estimated coefficient
                                                                                               1
corresponding to a dummy variable, such as 11, the percentage effect on price is 100(e1  1 ).
Moreover, given the natural log of the dependent variable, the predicted price is not a simple
sum of the variables multiplied by the coefficients. Equation (1) can be rewritten succinctly as
ln(Priceit ) = βX it +  it , where β is a vector of coefficients and X it is a conformable vector of
                                                                       2
explanatory variables. The expected price is E[ Price]  eβXit  0.5 , where 2 is the variance of
the error term. Thus, the predicted implicit price for a product with characteristics X1 rather
                       2             2
than X0 is: eβX1  0.5  eβX0  0.5 .

                                                               Results

Estimates of the semi-log hedonic price models are shown in table 4. Test results suggest the
presence of heteroskedasticity; consequently, we report White’s heteroskedasticity-consistent
standard errors.4 The models fit the data well, explaining 50%, 70%, and 66% of the egg price
variation in the United States, San Francisco/Oakland, and Dallas/Fort Worth markets,
respectively.
   As a first step in the analysis, we sought to determine whether the implicit prices were
identical across locations. To investigate this issue, a pooled model (or restricted model)
was estimated; data from all locations were combined and parameters were constrained to

  4
     Because of the nature of egg products, some variables are correlated. For example, production method and the color of eggs
were highly collinear (but not perfectly so) due to the fact that different egg-laying breeds tend to be used for specialty egg
production. To address this issue, we estimated the hedonic model sequentially, beginning with the simplest model with production
method only, and then estimation of the more complicated model with additional attribute variables. F-tests were carried out to test
for the significance of additional variables; results show all additional sets of variables are significant. Moreover, statistical tests do
not support dropping the variables related to egg color from the regression, and the estimates related to production method are not
particularly sensitive to inclusion of the color variables.
414 December 2010                               Journal of Agricultural and Resource Economics

Table 4. Semi-Log Hedonic Model Estimates
 Variable                       United States     San Francisco / Oakland   Dallas / Ft. Worth
 Constant                         −2.043***               −0.095***             −1.939***
                                  (0.039)                 (0.013)               (0.031)
 Production Method:
   Conventional Fresh               Base                    Base                  Base
   Cage-Free                       0.449***                0.195***               0.407***
                                  (0.010)                 (0.008)                (0.010)
   Cage-Free & Omega-3             0.209***                0.572***                 —
                                  (0.020)                 (0.009)
   Cage-Free & Organic             0.572***                0.379***               0.606***
                                  (0.014)                 (0.011)                (0.018)
   Fertile                         0.524***               −0.020**                  —
                                  (0.015)                 (0.006)
   Free-Range                      0.629***                  —                      —
                                  (0.054)
   Free-Range & Organic            0.530***                  —                      —
                                  (0.044)
   Natural                         0.206***                 0.013                0.065***
                                  (0.014)                  (0.018)              (0.011)
   Omega-3                         0.493***                 0.261***             0.357***
                                  (0.013)                  (0.016)              (0.011)
   Organic                         0.614***                 0.260***             0.590***
                                  (0.009)                  (0.006)              (0.019)
   Organic & Omega-3               0.709***                 0.255***             0.903***
                                  (0.016)                  (0.012)              (0.011)
   Pasteurized                     0.183***                 0.414***             0.986***
                                  (0.035)                  (0.012)              (0.010)
   Vegan Fed                       0.428***                 0.186***             0.522***
                                  (0.013)                  (0.006)              (0.009)
   None Indicated                 −0.026***                 0.025**             −0.030**
                                  (0.007)                  (0.009)              (0.011)
 Color:
  White                              Base                    Base                 Base
  Brown                             0.233***                0.150***              0.227***
                                   (0.006)                 (0.010)               (0.024)
   None Indicated                   0.033***                0.045**               0.024
                                   (0.011)                 (0.014)               (0.023)
 Size:
   Small                             Base                   Base                  Base
   Medium                           0.167***              −0.976***               0.174***
                                   (0.019)                (0.014)                (0.027)
   Large                            0.413***              −0.838***               0.422***
                                   (0.018)                (0.015)                (0.011)
   Extra Large                      0.481***              −0.768***               0.466***
                                   (0.019)                (0.010)                (0.012)
   Jumbo                            0.530***              −0.679***               0.475***
                                   (0.019)                (0.010)                (0.012)
   Super Jumbo                      0.839***                 —                      —
                                   (0.026)
   None Indicated                   0.488***                 —                    0.478***
                                   (0.031)                                       (0.019)

                                                                               ( continued . . . )
Chang, Lusk, and Norwood                    A Hedonic Analysis of Retail Egg Prices 415

Table 4. Continued
 Variable                   United States   San Francisco / Oakland   Dallas / Ft. Worth

 Package:
   Others                       Base                  Base                  Base
   Carton                     −0.032***             −0.057                −0.037**
                              (0.011)               (0.053)               (0.019)
   Cardboard Carton           −0.017                −0.062**                0.008
                              (0.012)               (0.021)                (0.016)
   Plastic Carton              0.005                −0.072**                0.018
                              (0.013)               (0.031)               (0.038)
   Styrofoam Carton           −0.129***              0.044***             −0.134***
                              (0.012)               (0.001)               (0.001)
   Paper Carton               −0.052*                0.008***             −0.003***
                              (0.027)               (0.001)               (0.000)
   Plastic Wrapped Carton      0.017                 0.028***             −0.055***
                              (0.018)               (0.005)               (0.006)
   Cardboard Box              −0.128***             −0.066***               0.169***
                              (0.020)               (0.008)               (0.011)
   None Indicated             −0.063***                —                    0.614***
                              (0.022)                                     (0.012)
 Brand Label:
   Specific                     Base                  Base                  Base
   Private                    −0.006                −0.103***             −0.072***
                              (0.005)               (0.009)               (0.011)
   #Eggs/Package              −0.030***             −0.035***             −0.024**
                              (0.003)               (0.009)               (0.012)
   (#Eggs/Package) 2           0.0004***             0.0004                0.0003
                              (0.0001)              (0.009)               (0.012)
 Quarterly Dummy:
   October–December             Base                   —                     —
   January–March               0.040***                —                     —
                              (0.006)
   April–June                 −0.033***                —                     —
                              (0.006)
   July–September             −0.057***                —                     —
                              (0.006)
 Monthly Dummy:
   December                       —                   Base                  Base
   January                        —                 −0.019**              −0.073***
                                                    (0.009)               (0.011)
   February                       —                 −0.025**              −0.043***
                                                    (0.008)               (0.011)
   March                          —                 −0.016**              −0.062***
                                                    (0.008)               (0.011)
   April                          —                 −0.040***             −0.144***
                                                    (0.008)               (0.011)
   May                            —                 −0.041***             −0.153***
                                                    (0.008)               (0.011)
   June                           —                 −0.056***             −0.206***
                                                    (0.008)               (0.011)

                                                                        ( continued . . . )
416 December 2010                                                        Journal of Agricultural and Resource Economics

Table 4. Continued
 Variable                                            United States           San Francisco / Oakland          Dallas / Ft. Worth

 Monthly Dummy (cont’d.):
  July                                                      —                          −0.057***                   −0.165***
                                                                                       (0.001)                     (0.001)
      August                                                —                          −0.052***                   −0.152***
                                                                                       (0.001)                     (0.001)
      September                                             —                          −0.031***                   −0.100***
                                                                                       (0.001)                     (0.001)
      October                                               —                          −0.036***                   −0.087***
                                                                                       (0.001)                     (0.001)
      November                                              —                          −0.028***                   −0.054***
                                                                                       (0.001)                     (0.001)
 R2                                                       0.503                          0.704                        0.664
 No. of Observations                                      28,701                         9,053                       11,158

 Notes: Single, double, and triple asterisks (*,**,***) denote statistical significance at the 10%, 5%, and 1% levels,
 respectively. Numbers in parentheses are White’s heteroskedastic-consistent standard errors.

equal.5 The hypothesis of equality of coefficients on the three data sets is strongly rejected
according to a Chow test (p < 0.01). There are many valid reasons to suspect the estimates
may differ by location, and this intuition is borne out by statistical tests.
   Given the rejection of the hypothesis that data from San Francisco and Dallas can be
pooled into a single hedonic price function, one might question the validity of the hedonic
model fit to the aggregate U.S. data (i.e., the first column of results in table 4). We report a
hedonic price function for the entire U.S. market because it is of interest to policy makers and
producers. The U.S. hedonic price function provides information on, for example, the differ-
ence in average price of cage-free vs. cage eggs in the United States, which is a statistic of
interest to many individuals. We interpret the U.S. model estimates as providing descriptive
statistics on the average prices of different types of eggs. It is important to note that the
hedonic estimates from a U.S. model will not necessarily equal egg price difference in any
particular location or market; moreover, the U.S. hedonic coefficients are not equal to the
average hedonic coefficients across all locations.
   Looking first at the U.S. market results (based on N = 28,701 quarterly price observations
from 1,985 SKUs observed over five years), table 4 shows that most estimated parameters
are statistically significant at the 0.05 level or lower. The coefficient corresponding to cage-
free is 0.449, which implies that the U.S. average price premium for cage-free eggs is
100  [exp (0.449) − 1] = 56.7% as compared to conventional egg prices. The U.S. average
price premium for cage-free eggs (56.7%) over conventional eggs is less than that for free-
range eggs (87.57%). Hens laying eggs labeled as both cage-free and free-range are kept
uncaged in barns, but typically only free-range hens have access to the outdoors. It should be
noted that federal standards for organic labeling require that hens be given outdoor access,
and thus organic could also be considered free-range.
   The Omega-3 coefficient for the overall U.S. model is 0.493, indicating an average price
premium of 63.7% for eggs advertised to have this health benefit. In the U.S. market, organic

  5
    The Dallas and San Francisco data represent such a small fraction of the overall U.S. data that any “double counting” in the
pooled model is quite minimal. In the first quarter of 2007, less than 4% of all U.S. egg sales in our data set are in the Dallas and
San Francisco markets.
Chang, Lusk, and Norwood                               A Hedonic Analysis of Retail Egg Prices 417

eggs receive about an 85% premium over conventional eggs. Bundling the attributes of
Omega-3 and organic generates a price premium of approximately 103%.
   The average U.S. price premium for brown eggs is about 26% compared to white eggs.
Differences in egg color stem from differences in breeds of hens laying the eggs. Although
shell color is unrelated to quality, nutrition, or taste, some people may perceive brown eggs as
healthier or of better quality than white ones. Because most cage-free and organic eggs are
brown, this suggests a significant portion of the observed price premium for these production
methods over conventional eggs in the market is likely attributable to color. One of the
reasons brown hens (and thus brown eggs) are used in cage-free and organic production
systems is that they are genetically predisposed to act more amicably toward other hens in the
open barn environment than the traditional breeds of white hens used in the cage system.
However, they also tend to be less productive. These productivity differences result in higher
production costs for brown eggs, which might also explain the higher price premium for
brown eggs over white.
   Results in table 4 also show how egg prices vary with other egg attributes such as egg size,
packaging material, branding, and package size. The results indicate that larger eggs generally
receive price premiums relative to smaller eggs. Prices are also influenced by packaging
material; eggs packaged in the most common types such as carton and styrofoam cartons tend
to be lower priced than those in other package types. The lowest-priced packaging materials
are the styrofoam carton and the cardboard box. Private-labeled eggs are priced about the
same as brand-labeled eggs, holding constant other egg characteristics. As the number of eggs
per package increases, the results show that the SKU price falls at a decreasing rate, perhaps
suggesting volume discounting.
   The last two columns in table 4 report the hedonic estimate models for the San Francisco
and Dallas regional egg markets. Results reveal that most explanatory variables are statis-
tically significant. The magnitudes of coefficients suggest Dallas is more similar to the United
States than it is to San Francisco. Surprisingly, all parameter estimates associated with
production method are larger in Dallas than in San Francisco. For example, the implied price
premium for organic eggs is about 80% in Dallas but only 30% in San Francisco. Likewise,
the price premium for cage-free eggs is about 50% in Dallas but only about 21% in San
Francisco. The lower percentage premiums in San Francisco vs. Dallas could be explained by
the fact that overall retail egg prices are higher in San Francisco than in Dallas. However, as
can be observed in table 3 (and as we show later in table 5), cage-free and organic price
premiums are higher in Dallas than in San Francisco, not only on a percentage basis but also
on an absolute dollar basis as well.
   Differences in price premiums across location may result from differences in demand. For
example, Dallas consumers may be willing to pay more for cage-free and organic eggs than
San Francisco consumers. However, supply differences could also be a factor. Specifically,
although consumers in San Francisco may have a higher willingness to pay for cage-free and
organic eggs than their Dallas counterparts, a potentially greater supply of these types of eggs
in California may result in a relatively lower market price. While it is plausible that supply
differences could explain the observed price premiums, it seems unlikely. Eggs can be
transported between California and Texas [the shelf life of an egg is around 3–5 weeks
(USDA/Food Safety and Inspection Service, 2010)], and California is a net importer of eggs
(Sumner et al., 2008).
   The highest priced egg in Dallas is the pasteurized egg—commanding a 168% premium
over conventional eggs. The implicit price for vegan fed (68.5% premium) is higher in Dallas
418 December 2010                                                       Journal of Agricultural and Resource Economics

than in San Francisco (20% premium). Implicit prices for “natural” eggs are close to zero in
Dallas and San Francisco. The effect of egg size is quite different between these two regional
markets. Whereas larger eggs enjoy a price premium in Dallas and in the rest of the United
States, small eggs command a price premium in San Francisco. It is unclear why this result
was obtained in San Francisco, but it stems from the sale of a single SKU that only appeared
in the last 44 weeks of the data set. As such, not much can (or should) be inferred from this
particular finding.
   At this point, it is worth asking whether the price premiums observed for cage-free and
free-range eggs can be attributed to consumers’ animal welfare concerns. First, the scientific
evidence clearly favors the notion that hen well-being is higher in cage-free production
systems than in the typical cage system [see the LayWel Research Project (2004) report, De
Mol et al. (2006), or Norwood and Lusk (2011)]. Consequently, there is a clear link between
these production systems and animal welfare. Still, it is possible that there are other unob-
served factors correlated with cage-free production. Might people buy cage-free eggs for other
reasons than production method used? Perhaps, but our regressions control for many of these
factors (color, package size, type of packaging, etc.).
   One obvious unobserved factor is brand name. We have information on individual brand
names for each SKU, but there are too many individual brands in our sample to separately
identify the effect of brand name from the individual product attributes and production
methods. We partially controlled for such factors with the Private vs. Specific brand variables
shown in table 4. However, even if we expand on this analysis by creating dummy variables
for each brand name and then omit any categorical variables that are subsequently
unidentified, we continue to find significant price premiums for cage-free eggs. For example,
applying this approach to the data from Dallas, we find that the coefficient on cage-free was
very similar to the result reported in table 4 (0.43 vs. 0.407), which does not control for each
individual brand name. We interpret such findings to imply that the estimated premium for
cage-free vs. cage eggs is unlikely to be caused by the presence of unmeasured brand-specific
characteristics. Of course, the cage-free label might carry additional connotations other than
improved animal well-being, such as perceptions of better taste. Accordingly, the observed
price premiums for “cage-free” include animal welfare characteristics in addition to issues
like taste and quality that consumers might infer from a cage-free label.
   Table 5 reports predicted price premiums for several types of eggs as compared to conven-
tional eggs on a dollar per dozen basis.6 The expected price for conventional white eggs was
about $1.77/dozen in the United States and $1.90/dozen in Dallas; the expected price in San
Francisco ($3.03/dozen) was substantially higher. The estimated price for a dozen cage-free,
brown eggs was about $3.50/dozen in the United States, $3.57/dozen in Dallas, and
$4.28/dozen in San Francisco. Thus, the price premiums for cage-free, brown eggs over
conventional, white eggs were about $1.70/dozen in the United States and Dallas but only
$1.25/dozen in San Francisco.
   Because cage-free and organic eggs are typically brown, whereas most conventional eggs
are typically white, it is of interest to ask how much of the premium for cage-free and organic
eggs typically observed is a result of egg color. To address this issue, table 5 compares the
predicted prices for both brown and white cage-free (and organic) eggs to the predicted price
for conventional, white eggs. For example, holding egg color constant at white, the U.S. data

  6
     Due to the nonlinear transformation of the regression error, the calculated price premiums are consistent but biased estimates
of the sample price premiums.
Chang, Lusk, and Norwood                                            A Hedonic Analysis of Retail Egg Prices 419

Table 5. Predicted Egg Prices and Premiums ($/dozen)
                                                                           San Francisco /             Dallas /
 Egg Type                                        United States                Oakland                 Ft. Worth
 Conventional, White                                 $1.77                     $3.03                    $1.90
 Cage-Free, White                                    $2.77                     $3.68                    $2.85
 Cage-Free, Brown                                    $3.50                     $4.28                    $3.57
 Organic, White                                      $3.27                     $3.93                    $4.68
 Organic, Brown                                      $4.12                     $4.57                    $5.87
 Cage-Free, White Premium a                          $1.00                     $0.65                    $0.95
 Cage-Free, Brown Premium a                          $1.73                     $1.25                    $1.68
 % Premium Attributable to Color b                  42.04%                    47.73%                   43.27%
 Organic, White Premium a                            $1.50                     $0.90                    $2.78
 Organic, Brown Premium a                            $2.36                     $1.54                    $3.97
 % Premium Attributable to Color b                  36.37%                    41.41%                   30.01%

 Note: Estimated values are for one dozen large eggs in a carton container in the fourth quarter of 2008.
 a
   Premium relative to conventional, white eggs.
 b
   Calculated as: [1 – (White Premium / Brown Premium)]  100.

set indicates cage-free eggs are predicted to command a premium of $1.00/dozen over con-
ventional eggs; however, because most cage-free eggs are brown, the price premium that will
typically be observed is instead $1.73/dozen. Based on these calculations, $0.73 of the premium
typically observed in a grocery store can be attributable to color, whereas the remaining $1.00
is a result of production practice (cage-free). Thus, 42% of the total premium typically
observed for cage-free eggs is linked to egg color, not the fact that hens were housed in cage-
free systems. Similarly, about 36.4% of the “typical” organic premium can be attributed to
egg color rather than organic production per se.

On the Demand for Cage-Free and Organic Eggs

To this point, we have focused on the descriptive models explaining how egg prices vary with
product characteristics. Table 4 provides estimates of the first-stage hedonic analysis and
reports the implicit prices of several egg characteristics. While they relate to a point on the
demand curve, they do not identify the demand curve as distinct from supply-side factors.
Hence, it is worth asking whether and to what extent our data can be used to provide structural
estimates of demand.
    In classic demand theory, consumer utility is defined over the quantity of goods consumed.
However, consumer utility can instead be written as a function of the characteristics or
attributes of the goods (e.g., Lancaster, 1966; Rosen, 1974; Gorman, 1980). It is assumed a
consumer will choose the good (which is defined by its underlying attributes or character-
istics) that maximizes utility, given prices and a budget constraint. The first-order condition
of this maximization problem indicates that the marginal price of an attribute (i.e., the
derivative of the price of the composite good with respect to the underlying attribute of
interest) will equal the consumer’s marginal rate of substitution between the attribute and a
numeraire (see Rosen, 1974; Taylor, 2003). As demonstrated by Rosen, one can rewrite the
optimization problem to derive an optimal bid or willingness-to-pay function. The result of this
optimization problem shows that the consumer’s marginal willingness to pay for an attribute is
420 December 2010                                                         Journal of Agricultural and Resource Economics

equal to the marginal rate of substitution between the attribute and the numeraire. Thus, from
these two equalities, we find that the marginal price of an attribute equals the consumer’s
marginal willingness to pay for the attribute (see Rosen). This insight is often used to link
implicit prices from a hedonic model like that shown in table 4 to consumers’ willingness to
pay at a point on the demand curve.
   For discrete characteristics, however, equality between the marginal implicit price and
marginal rate of substitution breaks down because one cannot differentiate the utility function
with respect to the discrete variable. The “1st stage” coefficients associated with discrete
characteristics provide an estimate of implicit prices, but are not technically marginal rates of
substitution (or willingness to pay). Yet, Bajari and Benkard (2005) argued that the coeffi-
cients associated with discrete attributes provide information or bounds on willingness to pay
given the existence of a price function and an assumption about the functional form for utility
(see also Bajari and Kahn, 2005). They went on to show that for discrete characteristics, such
as cage-free or organic, the implicit hedonic prices from the first-stage regression can be
coupled with data on the choices actually made by consumers to infer willingness to pay.
   In particular, the regression coefficients imply that consumers in Dallas faced an expected
implicit price premium of $0.95/dozen for cage-free eggs, holding color constant at white (see
table 5). It stands to reason that consumers who bought cage-free eggs in Dallas must have
been willing to pay at least $0.95/dozen for cage-free eggs, and consumers who chose not to
buy cage-free eggs must have had a willingness to pay of something less than $0.95/dozen for
this characteristic. Despite this high implicit price (or perhaps because of it), just 0.50% of
eggs purchased in Dallas were cage-free only (see table 2). A Turnbull lower-bound estimate
on willingness to pay for cage-free eggs in Dallas is 0.005  0.95 = $0.01/dozen (see Haab
and McConnell, 2002, p. 76). Similar reasoning suggests lower-bound estimates of about one
to three cents on cage-free and organic eggs in Dallas and San Francisco.
   One could take this line of reasoning further and actually estimate the mean willingness to
pay in the population using the approach outlined in Bajari and Kahn (2005). The approach
requires estimating the implicit price premium for cage-free or organic eggs at each data
point, and calculating the probability of purchasing cage-free at each SKU using an assumed
parametric function for the distribution of willingness to pay. Coupling this probability
statement with the actual quantity of each SKU sold in a likelihood function, one can arrive at
an estimate of the mean willingness to pay.
   When we apply this approach to our data, we find the estimated mean willingness-to-pay
values for cage-free and organic that rationalize the consumption data are very small.7 This
finding might seem counterintuitive initially, but it is perfectly consistent with the underlying
data. Although the implicit prices for organic or cage-free are quite high, there are few people
actually willing to pay the implicit price for cage-free and organic eggs, as evidenced by the
small market shares for these products (see table 2). While some people are willing to pay the
implicit price for cage-free or organic, most are not. Thus, in a population of shoppers, the
mean willingness to pay for cage-free (or organic) eggs is substantially less than the estimated
implicit price premium for cage-free (or organic) despite the existence of a few shoppers who
are willing to pay significant premiums for these products.

  7
    The unconstrained means are in fact negative. One could force the estimated willingness to pay for cage-free or organic to be
positive by picking a particular distribution such as a lognormal distribution or a uniform with a limit at zero; but in such cases, we
find that the likelihood function simply converges at the boundary, $0.
Chang, Lusk, and Norwood                               A Hedonic Analysis of Retail Egg Prices 421

                                         Conclusions

The U.S. fresh egg market has recently witnessed an explosion in product differentiation.
Using retail scanner data collected from a representative sample of grocery stores in the
United States, this study investigated the value of several egg attributes using the hedonic
method.
   Specialty eggs commanded significant price premiums over conventional eggs. For example,
the average U.S. percentage price premium for organic and Omega-3 eggs is over 100%.
Other characteristics, such as egg color, package type, and egg size, were also significantly
related to egg price. We also studied hedonic egg prices in two specific metropolitan areas—
San Francisco/Oakland, California, and Dallas/Fort Worth, Texas—and found significant
differences across locations. Somewhat surprisingly, estimated price premiums for cage-free
and organic attributes appear lower in the state for which a ballot referendum to ban battery
cage production recently passed than in a state many would think would be less open to
animal welfare issues. Although egg prices were substantially higher in San Francisco, the
price premiums for cage-free, organic, and other alternative production methods tended to be
higher in Dallas.
   Despite the high price premiums observed for cage-free and organic eggs, our data also
reveal that the market shares for such products are very small. These data suggest average
consumer willingness to pay in the population of egg shoppers is much less than the estimated
price premiums. It should be noted that our data are not particularly informative in precisely
articulating the average willingness-to-pay value, in part because cage-free and organic eggs
are always so much higher priced than conventional eggs. One would need survey or experi-
mental data to better project what choices people would make were prices more similar.
   One interesting finding emerging from our analysis is that the retail price premiums
observed for cage-free and free-range eggs appear much larger than the estimated cost
differences at the farm level. For example, Sumner et al. (2008) estimate the costs of
production to be about $0.31/dozen higher in cage-free vs. cage production systems at the
farm level, but our estimates suggest the U.S. average retail price premium for cage-free eggs
is $1.00/dozen. The data reported here are potentially revealing in what they have to say
about retailers’ pricing strategies. Although the market shares for cage-free and organic eggs
are small, such products appear in almost every major grocery store chain, indicating the price
premiums are probably not a result of the offerings of a particular kind of store. The price of
any one egg characteristic is influenced by the cost and value of the good, but price could also
be strategically set to encourage the patronage of certain consumers, allowing the food retailer
to charge higher premiums on other goods. Whole turkeys at Thanksgiving are a prime
example. Food retailers are suspected of selling turkeys at a loss on Thanksgiving, providing
consumers with a large consumer surplus, which may increase the value of shopping at the
store and increase consumers’ willingness to pay for other items (DeGraba, 2006). It is
possible that stores in some areas may sell cage-free eggs at lower prices than other areas as a
strategy to influence store volume and price premiums for other goods. Just as turkey prices
decline at precisely the same time demand rises, it is possible—though perhaps not
probable—that higher demand for cage-free eggs in one location may translate into lower
prices.
   Also interesting is the mix of products offered in the market place. Apparently egg
producers often bundle attributes and redundantly label attributes (e.g., an organic product is
both cage-free and free-range, but some packages explicitly label all characteristics whereas
422 December 2010                                         Journal of Agricultural and Resource Economics

others do not). Two often bundled characteristics (due to the particulars of egg production)
are egg color and cage-free production. Because our data set contains a few “unusual” SKUs
which are either brown and cage-raised or are white and cage-free, we can estimate the
portion of the “typical” price premium witnessed on cage-free eggs that results solely due to
color rather than a change in production practice. Based on our estimates, approximately 42%
of the typical cage-free premium and about 36% of the organic premium can be attributed to
the fact that these products are typically brown.
   The hedonic approach employed in this paper is useful for identifying the implicit prices of
egg characteristics. Such information is valuable for producers to know whether it is
profitable to undertake costly investments to change production systems, and it is also
beneficial in judging the potential impacts on market prices which may result from state-wide
ballot initiatives that ban certain production practices. Future research will focus on exploring
approaches for estimating consumer demand (rather than implicit prices) for egg production
characteristics, identifying the determinants of demand for more “animal-friendly” products,
and exploring retailer pricing strategies.

                [Received September 2009; final revision received August 2010.]

                                              References

American Egg Board. “Egg Industry Facts Sheet.” Online. Available at http://www.aeb.org/egg_industry/
    industry_facts/fact_sheet.html. [Accessed August 10, 2009.]
Asselin, A. M. “Eggcentric Behavior-Consumer Characteristics that Demonstrate Greater Willingness to Pay
    for Functionality.” Amer. J. Agr. Econ. 87(2005):1339–1344.
Bajari, P., and C. L. Benkard. “Demand Estimation with Heterogeneous Consumers and Unobserved Product
    Characteristics: A Hedonic Approach.” J. Polit. Econ. 113(2005):1239–1276.
Bajari, P., and M. E. Kahn. “Estimating Housing Demand with and Application to Explaining Racial
    Segregation in Cities.” J. Bus. and Econ. Statis. 23(2005):20–33.
Baltzer, K. “Consumers’ Willingness to Pay for Food Quality—The Case of Eggs.” Food Econ. 1(2004):78–90.
Bennett, R. M. “Farm Animal Welfare and Food Policy.” Food Policy 22(1997):281–288.
DeGraba, P. “The Loss Leader Is a Turkey: Targeted Discounts from Multi-Product Competitors.” Internat.
    J. Indust. Org. 24(2006):613–628.
De Mol, R. M., W. G. P. Schouten, E. Evers, H. Drost, H. W. J. Houwers, and A. C. Smits. “A Computer
    Model for Welfare Assessment of Poultry Production Systems for Laying Hens.” Netherlands J. Agr. Sci.
    54(2006):157–168.
Diewert, E. “Hedonic Regressions: A Review of Some Unsolved Issues.” Paper presented at Conference on
    Research in Income and Wealth, National Bureau of Economic Research, Cambridge, MA, 2003. Online.
    Available at http://www.nber.org/CRIW/papers/diewert.pdf. [Last accessed May 1, 2010.]
Fearne, A., and D. Lavelle. “Perceptions of Food ‘Quality’ and the Power of Marketing Communication:
    Results of Consumer Research on a Branded-Egg Concept.” J. Product & Brand Mgmt. 5(1996):29–42.
Gorman, W. M. “A Possible Procedure for Analyzing Quality Differentials in the Egg Market.” Rev. Econ.
    Stud. 47(1980):843–856.
Haab, T. C., and K. E. McConnell. Valuing Environmental and Natural Resources: The Econometrics of
    Non-Market Valuation. Northampton, MA: Edward Elgar Publishing, 2002.
Halvorsen, R., and R. Palmquist. “The Interpretation of Dummy Variables in Semilogarithmic Equations.”
    Amer. Econ. Rev. 70(1980):474–475.
Humane Society of the United States. “Humane State Ranking.” Online. Available at http://www.humane
    society.org/news/press_releases/2010/02/humane_state_rankings_020810.html. [Accessed February 2010.]
Karipidis, P., E. Tsakiridou, N. Tabakis, and K. Mattas. “Hedonic Analysis of Retail Egg Prices.” J. Food
    Distrib. Res. 36(2005):68–73.
Chang, Lusk, and Norwood                                      A Hedonic Analysis of Retail Egg Prices 423

Lancaster, K. “A New Approach to Consumer Theory.” J. Polit. Econ. 74(1966):132–157.
LayWel Research Project. “Welfare Implications of Changes in Production Systems for Laying Hens.” Pub.
   No. SSPE-CT-2004-502315, Specific Targeted Research Project (STReP), FP6 European Research
   Programme, 2004.
Lusk, J. L. “The Effect of Proposition 2 on the Demand for Eggs in California.” J. Agr. and Food Indust.
   Org. 8(2010): Issue No. 1, Article 3.
McConnell, K., and I. E. Strand. “Hedonic Prices for Fish: Tuna Prices in Hawaii.” Amer. J. Agr. Econ.
   82(2000):133–144.
Norwood, F. B., and J. L. Lusk. Compassion by the Pound: The Economics of Farm Animal Welfare. Oxford,
   UK: Oxford University Press, 2011.
Oberholtzer, L., C. Greene, and E. Lopez. “Organic Poultry and Eggs Capture High Price Premiums and
   Growing Share of Specialty Markets.” USDA/Economic Research Service Outlook Report No. LDP-M-
   150-01, Washington, DC, 2006. Online. Available at http://www.ers.usda.gov/Publications/LDP/2006/
   12Dec/LDPM15001/ldpm15001.pdf. [Last accessed May 1, 2010.]
Oczkowski, E. “A Hedonic Price Function for Australian Premium Table Wine.” Austral. J. Agr. Econ.
   38(1994):93–110.
Putnam, J. J., and J. E. Allshouse. “Food Consumption, Prices, and Expenditures, 1970–97.” Statis. Bull. No.
   SB-965, USDA/Economic Research Service, Washington, DC, 1999.
Rosen, S. “Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition.” J. Polit.
   Econ. 82(1974):34–55.
Sumner, D. A., J. T. Rosen-Molina, W. A. Matthews, J. A. Mench, and K. R. Richter. “Economic Effects of
   Proposed Restrictions on Egg-Laying Hen Housing in California.” University of California Agricultural
   Issues Center Report, July 2008.
Taylor, L. O. “The Hedonic Method.” In A Primer on Nonmarket Valuation, eds., P. A. Champ, K. J. Boyle,
   and T. C. Brown, pp. 331–393. Norwell, MA: Kluwer Academic Publishers, 2003.
U.S. Department of Agriculture, Economic Research Service. “Food Availability: Spreadsheets, Eggs: Per-
   Capita Availability,” 2009. Online. Available at http://www.ers.usda.gov/data/foodconsumption/Spread
   sheets/eggs.xls. [Last accessed May 1, 2010.]
U.S. Department of Agriculture, Food Safety and Inspection Service. “Fact Sheets: Egg Products Prep-
   aration.” Online. Available at http://www.fsis.usda.gov/Fact_Sheets/Focus_On_Shell_Eggs/index.asp. [Last
   accessed May 1, 2010.]
Waugh, F. V. “Quality Factors Influencing Vegetable Prices.” J. Farm Econ. 19(1928):185–196.
You can also read